Total PDE activity was determined from the remaining lysis buffer after RNA purification or with no purification (control). The lysates were diluted to 1 µg total protein per assay. The PDE assays were performed using Amersham's cAMP-PDE SPA Assay kit (GE Healthcare Europe, Brøndby, Denmark) in a 40 µL reaction mixture containing 1 µM cAMP, 50 mM Tris-HCl pH 7.5, 6 mM CaCl2, 8.3 mM MgCl2, 1.7 mM EGTA, and 1% DMSO. The reactions were stopped after 0, 5, 10, 15, 30, and 60 min by the addition of 2.5 mM Isobutylmethylxanthine (IBMX) and 10 µM 2-[(5-methyl-2-phenyl-1,3-oxazol-4-yl) methyl]-N-[(6-methylpyridin-2-yl)methyl]- 1,2,3,4-tetrahydroisoquinoline-4-carboxamide (THIQ, compound 2 in Reference 9) (LEO Pharma, Ballerup, Denmark). The enzyme activity was calculated by fitting a linear regression to the linear part of the reaction curve after subtracting the nonspecific activity determined by the addition of 1.25 mM IBMX and 5 µM THIQ to the lysate.
Protein concentrations were determined by the modified Bradford method, except for the lysate from the Qiagen Allprep kit where the Abs280 + 260 were used as recommended by the manufacturer due to incompatibility with the Bradford reagent.
The LDH assay was based on the method of Haslam et al. (10) using 5 µL lysate per well. The reaction mixture contained 0.32 mM β-nicotinamide adenine dinucleotide, 32 mM lithium lactate, 566 µM iodonitrotetrazolium chloride, 76 µM phenazine methosulfate, and 65 mM Tris-HCl, pH 8.5. The absorbance at 492 nm was measured every minute for 40 min, and the reaction rate was determined from the linear part of the curve after background subtraction estimated by omitting lithium lactate from the reaction buffer.
The caspase-3 activity was determined using the Caspase-3 Fluorescence Assay Kit (Cayman Chemical Company, Ann Arbor, MI) according to the manufacturer's instructions, except for a reduction in the reaction volume from 200 µL to 10 µL as this leads to a significant increase in the signal-to-noise ratio. The activity was calculated as the reaction rate during a 1 h incubation at 37°C after subtraction of nonspecific activity as determined by the addition of the included caspase-3 inhibitor.Statistical analysis
We used a linear mixed-effect model with litter as a random effect for comparison of RNA yields and quality and genomic DNA contamination between groups while a linear fixed-effect model was used for comparison of enzymatic activities between groups. A P value ≤0.05 was considered statistically significant.Results and discussion
This study demonstrates that it is possible to isolate RNA comparable in amount and quality to RNA isolated from existing kits while preserving functional protein, similar to that isolated from fresh samples. We tested the RNA yield for the standard oligo-dT bead protocol and found no differences in the Cq values obtained for the Gapdh, Rpl32 and Inha genes (Figure 1A). The PCR efficiency (Figure 1B) of our modified protocol was comparable to that of the Qiagen RNeasy kit.
The main difference compared to the existing oligo-dT protocol is the replacement of the ionic detergent with a non-ionic detergent to better preserve proteins in their native state. Thus, the lysate still contained active RNase, but its activity was sufficiently reduced by lowering the temperature to 4°C during hybridization of mRNA to the oligo-dT beads (11). However, this was below the optimal temperature for that hybridization (12). Therefore, we compared the effect of increasing the hybridization time from 5 to 20 min at various salt concentrations, which influence the melting temperature and hence the hybridization rate (12). Longer hybridization times significantly improved the yield (Figure 2). Reducing the salt concentration below 500 mM gave lower yields, even after 20 min, suggesting the need for even longer hybridization times if the downstream protein assay is not compatible with a high salt concentration.
We performed an immunoprecipitation experiment (Figure 3) to test if 500 mM NaCl affected antibody capture efficiency. We found no differences with one antibody (anti-GAPDH), while the capture efficiency of the other tested antibody (anti-β-actin) was severely reduced in the high salt buffer. Therefore, assays more sensitive to salt concentration (e.g., co-immunoprecipitation) may require a lower salt concentration and increased hybridization time due to the reduced hybridization rate.